Auxiliary vehicle searching method, system and device and storage medium

Abstract

The invention discloses an auxiliary vehicle searching method, system and device and a storage medium, and relates to the technical field of vehicle positioning, and the method comprises the steps: collecting an image around a vehicle under the condition that a parking signal is detected, carrying out the text detection of the image, obtaining the information of at least one parking space, screening the information of the parking space at least based on the position and confidence, and obtaining a vehicle positioning result; the method comprises the steps of screening residual parking space information, determining a parking space type according to the residual parking space information after screening, adding image deformation information to the position of a parking space identifier in an image in the residual parking space information after screening based on the parking space type, performing text extraction on the image with the image deformation information, obtaining the parking space identifier, and sending the parking space identifier to user equipment. According to the invention, the image deformation information is added to the position of the parking space identifier in the image, the text extraction is carried out based on the image with the image deformation information to obtain the parking space identifier, and the parking space identifier is sent to the user equipment, so that the user determines the parking position based on the parking space identifier, and the vehicle searching efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of automobile positioning technology, and in particular to a method, system, device and storage medium for assisting in finding a vehicle. BACKGROUND

[0002] In order to provide sufficient parking space for vehicles, the structure of large parking lots is becoming more and more complex, which causes the problem of difficulty in finding a vehicle. When a user forgets the parking area or the parking number cannot determine the accurate position of the parked vehicle, the user usually controls the vehicle by using a key to remind the position. However, due to the interference of the structure of the parking lot on the key signal, or the user loses the direction and the distance between the user and the vehicle exceeds the coverage range of the key signal, the user cannot quickly find the vehicle, which reduces the efficiency of finding the vehicle. Therefore, how to improve the efficiency of finding the vehicle has become a problem to be solved. SUMMARY

[0003] In view of the above problems, the present application provides a method, system, device and storage medium for assisting in finding a vehicle to improve the efficiency of finding the vehicle. The specific scheme is as follows:

[0004] The first aspect of the present application provides a method for assisting in finding a vehicle, comprising:

[0005] In the case of detecting a parking signal, an image around the vehicle is collected;

[0006] Text detection is performed on the image to obtain at least one parking space information, the parking space information comprising: a position of a parking space identifier in the image and a confidence of the parking space identifier; and the parking space information is filtered based at least on the position and the confidence;

[0007] According to the remaining parking space information after the filtering, a parking space type is determined, the parking space type comprising a vertical parking space and a lateral parking space;

[0008] Based on the parking space type, the position of the parking space identifier in the image in the remaining parking space information after the filtering is added with image deformation information;

[0009] Text extraction is performed on the image with the image deformation information to obtain the parking space identifier;

[0010] The parking space identifier is sent to a user device.

[0011] In a possible implementation, the determination of the parking space type according to the remaining parking space information after the filtering comprises:

[0012] For each of the parking space information:

[0013] obtaining a horizontal coordinate and a vertical coordinate of a center point of the position in a preset coordinate system, wherein an origin of the preset coordinate system is a vehicle center point, a vertical axis of the preset coordinate system is a straight line determined by a vehicle head center point, the vehicle center point and a vehicle tail center point, and a horizontal axis of the preset coordinate system is a straight line passing through the vehicle center point and being perpendicular to the vertical axis;

[0014] in a case where a minimum distance between the horizontal coordinate and the vertical axis is not greater than a first preset threshold and a minimum distance between the vertical coordinate and the horizontal axis is greater than a second preset threshold, determining that the parking space type is the perpendicular parking space;

[0015] in a case where the minimum distance between the horizontal coordinate and the vertical axis is greater than a third preset threshold and the minimum distance between the vertical coordinate and the horizontal axis is not greater than a fourth preset threshold, determining that the parking space type is the side parking space.

[0016] In a possible implementation, the adding, based on the parking space type, of image deformation information to a position of the parking space identifier in the image in the remaining parking space information after screening comprises:

[0017] for each of the parking space information remaining after screening:

[0018] identifying a relative angle of each character of the parking space identifier in the parking space information relative to the vehicle;

[0019] respectively calculating an angle difference between each of the relative angles and a preset character angle corresponding to the parking space type, and establishing a corresponding relationship between the angle difference and the character, the angle difference being an angle vector with a deviation direction;

[0020] adding the angle difference and the corresponding relationship as the image deformation information to the position of the parking space identifier in the image.

[0021] In a possible implementation, the text extraction on the image with the image deformation information to obtain the parking space identifier comprises:

[0022] according to the angle difference in the image deformation information, performing rotation correction on each character of the parking space identifier in the position added with the image deformation information in the image, so that a relative angle of each character relative to the vehicle is equal to the preset character angle, and performing image extraction on the position after the rotation correction to obtain a parking space identifier image;

[0023] performing the text extraction on the parking space identifier image to obtain the parking space identifier.

[0024] In a possible implementation, the text detection on the image to obtain at least one parking space information comprises:

[0025] identify a position of each string image in the current parking space in the image, a minimum occupied area of the string image in the image, and a confidence of the string image, and determine the position, the minimum occupied area, and the confidence of the same string image as one piece of the parking space information.

[0026] In a possible implementation, the screening of the parking space information based at least on the position and the confidence includes:

[0027] eliminating other pieces of parking space information except for target parking space information, the target parking space information being: the confidence being in a preset confidence interval, the minimum occupied area having a size not less than a first preset size threshold, and the position being in a preset spatial region range, and the position being the closest to the center of the vehicle, and the preset spatial region range being a region range formed by multiple arrangement regions of parking space identifiers in the parking space.

[0028] In a possible implementation, the method further includes:

[0029] sending the image together with the parking space identifier to the user equipment.

[0030] In a possible implementation, the sending of the image together with the parking space identifier to the user equipment includes:

[0031] performing compression encoding processing on the image to obtain a compressed image data packet;

[0032] splitting the compressed image data packet into multiple image sub-packets, and performing format packaging on the parking space identifier and each image sub-packet; and sending the format-packaged parking space identifier and each image sub-packet to the user equipment.

[0033] In a possible implementation, the sending of the format-packaged parking space identifier and each image sub-packet to the user equipment includes:

[0034] establishing a secure data transmission channel with the user equipment;

[0035] sending the format-packaged parking space identifier to the user equipment through the secure data transmission channel, so that the user equipment analyzes and displays the format-packaged parking space identifier;

[0036] send the image packages encapsulated by the format to the user equipment one by one through the secret data transmission channel, and after all the image packages encapsulated by the format have been sent to the user equipment, send an end transmission instruction to the user equipment, so that the user equipment, in response to the end transmission instruction, parses and restores the image packages encapsulated by the format that have been received to obtain the image.

[0037] In a possible implementation, the sending of the image packages encapsulated by the format to the user equipment one by one through the secret data transmission channel comprises:

[0038] determining a sending order of the image packages encapsulated by the format according to the split order;

[0039] determining one of the image packages encapsulated by the format that has not been sent to the user equipment and has the most optimal sending order as a target package, and sending the target package to the user equipment;

[0040] after receiving the continue receiving instruction fed back by the user equipment, marking the target package as having been sent, and returning to the operation of determining one of the image packages encapsulated by the format that has not been sent to the user equipment and has the most optimal sending order as a target package, and sending the target package to the user equipment.

[0041] The second aspect of the application provides an auxiliary vehicle searching system, comprising:

[0042] an image acquisition module configured to acquire an image around a vehicle when a parking signal is detected;

[0043] a screening module configured to perform text detection on the image to obtain at least one parking space information, the parking space information comprising a position of a parking space identifier in the image and a confidence degree of the parking space identifier, and screen the parking space information based at least on the position and the confidence degree;

[0044] a parking space type determination module configured to determine a parking space type according to the parking space information remaining after screening, the parking space type comprising a vertical parking space and a side parking space;

[0045] an information adding module configured to add image deformation information to the position of the parking space identifier in the image in the parking space information remaining after screening based on the parking space type;

[0046] a text extraction module configured to perform text extraction on the image with the image deformation information to obtain the parking space identifier;

[0047] The first data sending module is configured to send the parking space identifier to a user equipment.

[0048] In a possible implementation, the parking space type determining module is configured to:

[0049] For each of the parking space information:

[0050] obtain a horizontal coordinate and a vertical coordinate of a center point of the position in a preset coordinate system, wherein an origin of the preset coordinate system is a vehicle center point, a vertical axis of the preset coordinate system is a straight line defined by a vehicle head center point, the vehicle center point and a vehicle tail center point, and a horizontal axis of the preset coordinate system is a straight line passing through the vehicle center point and being perpendicular to the vertical axis;

[0051] determine the parking space type as the perpendicular parking space when a minimum distance between the horizontal coordinate and the vertical axis is not greater than a first preset threshold and a minimum distance between the vertical coordinate and the horizontal axis is greater than a second preset threshold;

[0052] determine the parking space type as the side parking space when the minimum distance between the horizontal coordinate and the vertical axis is greater than a third preset threshold and the minimum distance between the vertical coordinate and the horizontal axis is not greater than a fourth preset threshold.

[0053] In a possible implementation, the information adding module is configured to:

[0054] For each of the parking space information remaining after the screening:

[0055] identify a relative angle of each character of the parking space identifier in the parking space information relative to the vehicle;

[0056] calculate an angle difference between each of the relative angles and a preset character angle corresponding to the parking space type, and establish a corresponding relationship between the angle difference and the character, wherein the angle difference is an angle vector with a deviation direction;

[0057] add the angle difference and the corresponding relationship to a position of the parking space identifier in the image as the image deformation information.

[0058] In a possible implementation, the text extracting module is configured to:

[0059] perform rotation correction on each character of the parking space identifier in the position added by the image deformation information in the image according to the angle difference in the image deformation information, so that a relative angle of each character relative to the vehicle is equal to the preset character angle, and perform image extraction on the position after the rotation correction to obtain a parking space identifier image.

[0060] The text is extracted from the parking space sign image to obtain the parking space sign.

[0061] In one possible implementation, the filtering module is configured to perform text detection on the image to obtain at least one parking space information as follows:

[0062] Identify the position of each string image in the current parking space, the minimum occupied area of ​​the string image, and the confidence level of the string image in the image, and determine the position, the minimum occupied area, and the confidence level of the same string image as a parking space information.

[0063] In one possible implementation, the filtering module is configured to filter the parking space information based at least on the location and the confidence level as follows:

[0064] Remove all parking space information except for the target parking space information. The target parking space information is the parking space information that has a confidence level within a preset confidence level range, a minimum occupied area size not less than a first preset size threshold, a location within a preset spatial area range, and a location with the smallest distance from the vehicle center. The preset spatial area range is the area range composed of multiple layout areas of parking space identifiers in the parking space.

[0065] In one possible implementation, the system further includes:

[0066] The second data transmission module is used to send the image along with the parking space identifier to the user equipment.

[0067] In one possible implementation, the second data sending module is specifically configured as follows:

[0068] The image is compressed and encoded to obtain a compressed image data packet;

[0069] The compressed image data packet is split into multiple image packets, and the parking space identifier and each of the image packets are format-encapsulated; the format-encapsulated parking space identifier and each format-encapsulated image packet are sent to the user equipment.

[0070] In one possible implementation, the second data sending module is configured to send the parking space identifiers and image packets, encapsulated in the aforementioned format, to the user equipment as follows:

[0071] Establish a secure data transmission channel with the user equipment;

[0072] The parking space identifier encapsulated in the specified format is sent to the user equipment through the secure data transmission channel, so that the user equipment can parse and display the parking space identifier encapsulated in the specified format.

[0073] Each image packet encapsulated in the specified format is sent to the user equipment one by one through the secure data transmission channel. After all image packets encapsulated in the specified format have been sent to the user equipment, a termination transmission command is sent to the user equipment so that the user equipment responds to the termination transmission command by parsing and restoring the received image packets encapsulated in the specified format to obtain the image.

[0074] In one possible implementation, the second data sending module is configured to send each image packet, encapsulated in the aforementioned format, to the user equipment one by one through the secure data transmission channel:

[0075] The order in which the image packets encapsulated in the specified format are sent is determined according to the order in which they are split.

[0076] The image packet that has not been sent to the user equipment and has the highest priority in the sending order is identified as the target packet after being encapsulated in the specified format, and the target packet is sent to the user equipment.

[0077] After receiving the continued receiving instruction from the user equipment, the target packet is marked as sent, and the process returns to execute the operation steps of determining the image packet encapsulated in the specified format that was not sent to the user equipment and has the highest priority in the sending order as the target packet, and sending the target packet to the user equipment.

[0078] A third aspect of this application provides an electronic device, comprising at least one processor and a memory connected to the processor, wherein:

[0079] The memory is used to store computer programs;

[0080] The processor is used to execute the computer program so that the electronic device can implement the vehicle-finding method of the first aspect or any implementation thereof.

[0081] A fourth aspect of this application provides a computer storage medium carrying one or more computer programs, which, when executed by an electronic device, enable the electronic device to implement the vehicle-finding method described in the first aspect or any implementation thereof.

[0082] By employing the aforementioned technical solution, this application provides an auxiliary vehicle location method, system, device, and storage medium. Upon detecting a parking signal, it acquires images of the area surrounding the vehicle and performs text detection on the images to obtain at least one parking space information, including the position of the parking space marker in the image and the confidence level of the marker. This achieves the acquisition of parking location information. Subsequently, by filtering the parking space information based on position and confidence level, non-parking space information is eliminated, improving the accuracy of subsequent parking space marker recognition. Furthermore, by determining the parking space type based on the remaining filtered parking space information, image distortion information is added to the position of the parking space marker in the image based on the parking space type. Text extraction is then performed on the image containing the image distortion information to obtain the parking space marker, avoiding the risk of reduced accuracy in parking space marker extraction due to image distortion. Finally, by sending the parking space marker to the user's device, the user can accurately determine their parking location based on the marker, thus improving vehicle location efficiency compared to existing methods where users memorize locations. Attached Figure Description

[0083] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0084] Figure 1 A flowchart of an auxiliary vehicle-finding method provided in this application;

[0085] Figure 2 A schematic diagram for determining the extraction area provided in this application;

[0086] Figure 3 A schematic diagram of a side parking space sign provided for this application;

[0087] Figure 4 A schematic diagram of a parking space sign for a perpendicular parking space provided in this application;

[0088] Figure 5 A schematic diagram of a preset coordinate system provided for this application;

[0089] Figure 6 A schematic diagram of a relative angle provided for this application;

[0090] Figure 7 A flowchart for sending an image along with a parking space identifier to a user device is provided in this application;

[0091] Figure 8 A flowchart illustrating a possible implementation of this application of an auxiliary vehicle-finding method;

[0092] Figure 9 A block diagram of an auxiliary vehicle finding system provided in this application;

[0093] Figure 10 This is a schematic diagram of the structure of an electronic device provided in this application. Detailed Implementation

[0094] The embodiments of this application are described below with reference to the accompanying drawings. The terminology used in the implementation section of this application is for explaining specific embodiments only and is not intended to limit the scope of this application.

[0095] The embodiments of this application will now be described with reference to the accompanying drawings. Those skilled in the art will recognize that, with technological advancements and the emergence of new scenarios, the technical solutions provided in the embodiments of this application are equally applicable to similar technical problems.

[0096] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms are interchangeable where appropriate; this is merely a way of distinguishing objects with the same attributes in the embodiments of this application. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, so that a process, method, system, product, or apparatus that comprises a series of elements is not necessarily limited to those elements, but may include other elements not explicitly listed or inherent to those processes, methods, products, or apparatuses.

[0097] The first aspect of this application provides an auxiliary vehicle location method, such as... Figure 1 As shown, this vehicle-finding assistance method includes:

[0098] S101. When a stop signal is detected, acquire images of the area around the vehicle.

[0099] It should be noted that in practical applications, the aforementioned parking signals are vehicle control signals used to indicate that the vehicle is about to stop and lose power, and can be at least one type. For example: shifting the vehicle gear from drive to brake, opening the parking brake, or closing all vehicle windows. This application, by configuring image acquisition and subsequent processing steps to begin upon detecting a parking signal, enables the transmission of parking space identification even when the vehicle is not powered off and the distance between the user and the vehicle does not exceed the communication distance between the vehicle and the user's device. This avoids the risk of the parking space identification failing to transmit due to the vehicle being powered off or the user being too far away, thus improving the reliability of vehicle location.

[0100] It should be noted that in practical applications, the images surrounding the vehicle mentioned above can be images recording the scene around the vehicle. These images can be captured by external cameras on the vehicle, such as Around View Monitor (AVM) images and fisheye images. Specifically, AVM images can be stitched together from images of the vehicle's undercarriage and perimeter.

[0101] It should be noted that in practical applications, there are many interfering objects around parking spaces, such as promotional posters scattered near the parking spaces, pedestrian warning information sprayed on the ground, and decorative text sprayed on other vehicles. These objects can interfere with the subsequent detection of parking space information and reduce detection efficiency. Therefore, to avoid the impact of interfering objects around the parking spaces on the accuracy of subsequent parking space identification extraction, the image around the vehicle can be cropped before performing step S102 to reduce the impact of interfering objects on subsequent processing steps. Specific operation steps may include steps A1 to A3 as follows.

[0102] Step A1: Input the image around the vehicle into the preset region of interest extraction algorithm and trigger step A2.

[0103] Step A2: A preset region of interest extraction algorithm determines the smallest bounding box containing the vehicle from the image surrounding the vehicle. This triggers step A3.

[0104] Step A3: Extend each side of the smallest rectangle outward by a preset distance, and determine the rectangular area formed by the extension as the extraction area, and extract it to obtain an image for text detection.

[0105] The above-mentioned diagram illustrating the determination of the extraction region is shown below. Figure 2 As shown, the black box containing the vehicle in the image is the smallest rectangular box in step A2 above. The ROI map is a rectangular region formed by extension. The AVM surround view is the image around the vehicle input in step A1 using the preset region of interest extraction algorithm.

[0106] S102. Perform text detection on the image to obtain at least one parking space information, including: the position of the parking space sign in the image and the confidence level of the parking space sign; filter the parking space information based at least on the position and confidence level.

[0107] It should be noted that in practical applications, there are multiple ways to implement text detection in images. Here, one example is provided:

[0108] The image is input into a preset deep learning algorithm for text detection, which detects and recognizes characters in the image, determines the position of each character in the image, and assigns confidence that the character represents a parking space sign. This preset deep learning algorithm can be of various types, such as the Paddle Optical Character Recognition (PaddleOCR) model, Fully Convolutional Networks (FCN) model, or Connectionist Text Proposal Network (CTPN) algorithm from a text model suite.

[0109] The training process of the aforementioned pre-defined deep learning algorithm for text detection can be as follows:

[0110] Obtain multiple images of the area around vehicles with different parking space types. The images include parking space markings and interference characters. Add location labels and confidence labels to each parking space marking in each image to obtain a training set.

[0111] An initial deep learning algorithm for text detection was trained using images of vehicles surrounding the training set with added location and confidence labels. Based on the location, confidence score, location label, and confidence label output by the initial deep learning algorithm, parameter tuning was performed to obtain the aforementioned preset deep learning algorithm for text detection. The input to this preset deep learning algorithm is images of the areas surrounding the vehicles, and the output is information on multiple parking spaces within those images. This parking space information includes the location of the parking space label in the image and the confidence score that the label is a valid parking space label.

[0112] It should be noted that in practical applications, parking space signs differ from interfering objects due to their form and relative position to vehicles. Therefore, this application configures text detection on the image to obtain parking space information, including the position of the parking space sign in the image and the confidence level of the parking space sign. Based on the confidence level and position, the parking space information is filtered, thereby eliminating interfering objects in the image and improving the accuracy of subsequent parking space sign extraction.

[0113] S103. Based on the remaining parking space information after filtering, determine the parking space type, which includes perpendicular parking spaces and lateral parking spaces.

[0114] It should be noted that in practical applications, the location of parking space markings varies depending on the type of parking space to facilitate users recording their parking location. For example, the markings for parallel parking spaces are often painted on the side of the car door. Figure 3As shown, parking space markings for perpendicular parking spaces are mostly painted on the front or rear side of the vehicle, such as... Figure 4 As shown. However, because the types of external cameras on vehicles differ at different locations—for example, wide-angle cameras are often used at the front or rear of the vehicle, resulting in lower image distortion—cameras located at the rearview mirror position often use fisheye cameras to ensure a wider field of view, leading to higher image distortion. Therefore, this application configures the system to determine the parking space type based on the remaining parking space information after filtering, thereby correcting distorted parking space signs during subsequent parking space sign extraction and improving the accuracy of parking space sign recognition and extraction.

[0115] S104. Based on the parking space type, add image deformation information to the position of the parking space identifier in the image of the remaining parking space information after filtering.

[0116] It should be noted that in practical applications, due to image distortion and parking space sign angle shifts, the parking space signs in the image are not arranged according to everyday reading and writing habits (horizontally arranged from left to right). Therefore, this application adds image deformation information to the position of the parking space signs in the image based on the parking space type. In the subsequent text extraction process, the text extracted from the corresponding position is corrected according to the image deformation information, thereby obtaining parking space signs that meet everyday reading and writing habits. This avoids the risk of recognition errors caused by parking space signs not meeting everyday reading and writing habits when users identify them by themselves based on the image, and improves the efficiency of finding the car.

[0117] S105. Extract text from the image containing image deformation information to obtain parking space markings.

[0118] It should be noted that in practical application scenarios, there are multiple implementation methods for extracting text from images with image deformation information to obtain parking space markings. Here, one is provided as an example, including the following steps B1 to B3.

[0119] Step B1: Invoke the preset correction script to correct the character image of the parking space sign in the location where image deformation information has been added to the image. Then trigger step B2.

[0120] Step B2 involves extracting the corrected character image from the image to obtain the parking space sign image, which then triggers step B3.

[0121] Step B3: Input the parking space sign image into a preset text recognition deep learning model for text extraction to obtain the parking space sign.

[0122] It should be noted that in practical applications, the preset text recognition deep learning model in step B3 above can be of various types, such as Convolutional Recurrent Neural Network (CRNN), Recurrent Neural Network (RNN), and Long Short-Term Memory (LSTM). When training the preset text recognition deep learning model, training data can be obtained by adding labels containing the correct characters of the text in the image to the text image. This training data is then used to train the initial text recognition deep learning model. Based on the labels and the recognition results of the initial text recognition deep learning model, the parameters of the initial model are tuned to obtain the preset text recognition deep learning model. After training, the input to the preset text recognition deep learning model is a parking space sign image, and the output is the characters of the parking space sign.

[0123] S106. Send the parking space identifier to the user equipment.

[0124] It should be noted that in practical application scenarios, there are multiple ways to send the parking space identifier to the user device. Here, one example is provided, which includes the following steps C1 to C3.

[0125] Step C1: Upon detecting a parking signal, a broadcast packet is sent to enable the digital vehicle key to establish a secure data transmission channel with the vehicle's infotainment system based on the broadcast packet. This triggers step C2.

[0126] Step C2: After obtaining the parking space identifier, the parking space identifier is encapsulated based on a preset data transmission protocol to obtain packet data. This triggers step C3.

[0127] Step C3: Send the packet data to the user equipment equipped with the digital vehicle key through the secure data transmission channel in step C1.

[0128] It should be noted that the digital vehicle key in steps C1 and C3 above is vehicle key software based on the CarConnectivity Consortium (CCC) protocol. Because the CCC protocol features high confidentiality and high transmission strength, transmitting parking space identifiers based on the CCC protocol can improve the accuracy and reliability of parking space identifier transmission in complex parking lots with high pedestrian traffic.

[0129] This application acquires parking location information by acquiring images of the area surrounding the vehicle upon detecting a parking signal and performing text detection on the images. This yields at least one parking space, including the location and confidence level of the parking space marker within the image. Subsequently, by filtering the parking space information based on location and confidence level, non-parking space information is eliminated, improving the accuracy of subsequent parking space marker recognition. Furthermore, by determining the parking space type based on the remaining filtered parking space information, image distortion information is added to the location of the parking space marker in the image based on the parking space type. Text extraction is then performed on the image containing the image distortion information to obtain the parking space marker, avoiding the risk of reduced accuracy due to image distortion. Finally, by sending the parking space marker to the user's device, the user can accurately determine their parking location based on the marker, improving parking location efficiency compared to existing methods where users memorize locations. Therefore, this application significantly improves parking location efficiency.

[0130] In one possible implementation, the parking space type is determined based on the remaining parking space information after filtering, including:

[0131] Information on each parking space:

[0132] Obtain the x and y coordinates of the center point of the location in a preset coordinate system. The origin of the preset coordinate system is the center point of the vehicle. The y-axis of the preset coordinate system is the straight line determined by the center point of the front of the vehicle, the center point of the vehicle, and the center point of the rear of the vehicle. The x-axis of the preset coordinate system is the straight line that passes through the center point of the vehicle and is perpendicular to the y-axis.

[0133] If the minimum distance between the horizontal axis and the vertical axis is not greater than the first preset threshold, and the minimum distance between the vertical axis and the horizontal axis is greater than the second preset threshold, the parking space type is determined to be a perpendicular parking space.

[0134] If the minimum distance between the horizontal axis and the vertical axis is greater than the third preset threshold, and the minimum distance between the vertical axis and the horizontal axis is not greater than the fourth preset threshold, the parking space type is determined to be a side parking space.

[0135] To facilitate understanding of the above-described process for determining parking space types, an example of one possible implementation of this application is provided below:

[0136] like Figure 5 The diagram shows a schematic of a preset coordinate system. The origin of the preset coordinate system is the center point of the vehicle. Rectangle 51 represents the parking space, rectangle 52 represents the position of the parking space indicator when the parking space is a perpendicular parking space, and rectangle 54 represents the position of the parking space indicator when the parking space is a lateral parking space.

[0137] The x-coordinate of the center point 53 of the rectangle 52 lies on the y-axis of the preset coordinate system, meaning the minimum distance from the center point 53 to the y-axis is 0, and the minimum distance from the y-coordinate of the center point 53 to the x-axis is Y1. Therefore, if Y1 is greater than the second preset threshold, the current parking space can be determined to be a perpendicular parking space.

[0138] The ordinate of the center point 55 of the rectangle 54 lies on the horizontal axis X of the preset coordinate system, meaning the minimum distance from the center point 55 to the horizontal axis is 0, and the minimum distance from the horizontal coordinate of the center point 55 to the vertical axis Y is X1. Therefore, if X1 is greater than the third preset threshold, the current parking space can be determined as a parallel parking space.

[0139] In one possible implementation, the first preset threshold, the second preset threshold, the third preset threshold, and the fourth preset threshold can be set according to the parking space parameters and the length and width of the vehicle. For example, the first preset threshold can be set to 1 / 2 or 1 / 3 of the vehicle width, the second preset threshold can be set to 1 / 3 of the vehicle length, the third preset threshold can be set to 1 / 3 of the vehicle width, and the fourth preset threshold can be set to 1 / 3 or 1 / 4 of the vehicle length.

[0140] In one possible implementation, based on the parking space type, image deformation information is added to the position of the parking space identifier in the image among the remaining filtered parking space information, including:

[0141] For the remaining parking spaces after filtering:

[0142] The relative angle of each character in the parking space identifier relative to the vehicle is identified in the parking space information.

[0143] Calculate the angle difference between each relative angle and the preset character angle corresponding to the parking space type, and establish the correspondence between the angle difference and the character. The angle difference is an angle vector with a deviation direction.

[0144] The angle difference and its correspondence are used as image deformation information and added to the position of the parking space sign in the image.

[0145] It should be noted that, in practical applications, the relative angles of the characters to the vehicle refer to the angle between the characters' arrangement according to everyday reading habits and the line connecting the center of the front of the parked vehicle and the center of the vehicle. For example: Figure 6 The diagram shown is a schematic representation of a relative angle. It can be seen that the characters in the parking space signs, arranged according to everyday reading habits, are oriented in the following direction. Figure 6The direction indicated by the middle arrow clearly shows that the relative angle between the character and the vehicle is 0 degrees. Since this parking space is a parallel parking space, the preset character angle for parallel parking spaces is set to 90 degrees. Therefore, the angle difference between the relative angle of the parking space sign character and the preset angle is 90 degrees, and the direction of this angle difference is the corrected character arrangement direction, horizontally aligned from left to right relative to the image.

[0146] In one possible implementation, text extraction is performed on the image containing image deformation information to obtain parking space identification, including:

[0147] Based on the angle difference in the image deformation information, the characters of the parking space sign at the location where the image deformation information is added are rotated and corrected so that the relative angle of each character to the vehicle is equal to the preset character angle. The image of the rotated and corrected position is then extracted to obtain the parking space sign image.

[0148] Text extraction is performed on the parking space sign image to obtain the parking space sign.

[0149] It should be noted that in practical application scenarios, this application calculates the relative angle of the characters of the parking space sign to the vehicle and the angle difference between the characters and the preset character angle corresponding to the parking space type. Based on the angle difference, the characters are rotated and corrected so that the parking space sign in the obtained parking space sign image after rotation correction meets the daily reading and writing habits. Therefore, users do not need to identify the parking space sign in the image by rotating their mobile phones or other means, which improves the user experience and the efficiency of finding the car.

[0150] In one possible implementation, text detection is performed on the image to obtain at least one parking space information, including:

[0151] Identify the position of each string image in the current parking space, the minimum occupied area of ​​the string image, and the confidence level of the string image in the image, and determine the position, minimum occupied area, and confidence level of the same string image as a parking space information.

[0152] It should be noted that in practical applications, the minimum occupied area of ​​the aforementioned string image in the image is the smallest rectangular area of ​​multiple consecutively arranged characters in the image. Since the position and size of the parking space sign in the parking space differ from interfering objects, this application identifies the position and minimum occupied area of ​​each string image in the current parking space, and determines the confidence level of each string image as a parking space sign. Through subsequent filtering, the string image with the highest probability of being a parking space sign is selected from multiple string images, thus eliminating interfering objects and improving the final recognition accuracy of the parking space sign.

[0153] In one possible implementation, parking space information is filtered based at least on location and confidence level, including:

[0154] Remove all parking space information except for the target parking space information. The target parking space information is: the parking space information whose confidence level is within the preset confidence level range, whose minimum occupied area size is not less than the first preset size threshold, whose location is within the preset spatial area range, and whose location is the smallest distance from the center of the vehicle. The preset spatial area range is the area range composed of multiple layout areas of the parking space identifier in the parking space.

[0155] It should be noted that, in practical applications, the aforementioned first preset size threshold and the aforementioned preset confidence interval can be set by analyzing the size and recognizability of parking space signs in different parking lots. This application does not impose excessive limitations on the specific parameters of the aforementioned preset confidence interval and the aforementioned first preset size threshold.

[0156] In one possible implementation, the above is as follows: Figure 1 The auxiliary vehicle location methods shown also include:

[0157] The image is sent to the user's device along with the parking space sign.

[0158] It should be noted that in practical application scenarios, if the parking space sign cannot be accurately identified and extracted due to reasons such as dirt, water accumulation, wear and tear (e.g., a certain character is missing), this application sends the image along with the parking space sign to the user's device, so that the user can compare the image with the image around the parking location, thereby assisting in finding the car. Compared with the method of blindly searching for the car manually, this improves the efficiency of finding the car.

[0159] In one possible implementation, sending the image along with the parking space identifier to the user equipment includes:

[0160] The image is compressed and encoded to obtain a compressed image data packet;

[0161] The compressed image data packet is split into multiple image packets, and the parking space identifier and each image packet are format-encapsulated; the format-encapsulated parking space identifier and each format-encapsulated image packet are sent to the user equipment.

[0162] It should be noted that in practical applications, the aforementioned compression encoding is an image compression method. By transforming, compressing, and encoding image data while meeting preset fidelity requirements, it reduces redundant data and the amount of data displayed in the image. This application reduces the amount of image data and improves the transmission efficiency of sending images to user devices by configuring image compression encoding processing.

[0163] It should be noted that, in practical applications, the aforementioned method of packetizing compressed image data involves splitting the compressed image data packet, which has been compressed and encoded into binary data, into multiple data segments, each of which constitutes an image packet. This application reduces the amount of data transmitted in a single transmission and improves data transmission efficiency by splitting the compressed image data packet into multiple image packets.

[0164] In one possible implementation, the formatted parking space identifiers and individual image packets are sent to the user equipment, including:

[0165] Establish a secure data transmission channel with user equipment;

[0166] The formatted parking space identifiers are sent to the user equipment through a secure data transmission channel, so that the user equipment can parse and display the formatted parking space identifiers.

[0167] Each formatted image packet is sent to the user equipment one by one through a secure data transmission channel. After all formatted image packets have been sent to the user equipment, a termination command is sent to the user equipment so that the user equipment can respond to the termination command by parsing and restoring the received formatted image packets to obtain the image.

[0168] It should be noted that in practical application scenarios, there can be multiple ways to establish a secure data transmission channel with the user equipment. Here, one example is provided, which includes the following steps D1 to D2.

[0169] Step D1: Upon detecting a stop signal, the vehicle's infotainment system sends an encrypted broadcast packet, triggering step D2.

[0170] In step D2, the digital vehicle key deployed in the user equipment captures the encrypted broadcast packet and decrypts it using its own configured decryption key; and if the decryption is successful, a secure data transmission channel with the vehicle system is established based on the interface address in the encrypted broadcast packet.

[0171] This application establishes a secure data transmission channel with the user equipment by configuring it, which avoids sending images to the wrong user equipment and also avoids interference from interference sources in the image transmission process, thereby improving image transmission efficiency.

[0172] In one possible implementation, the format-encapsulated image packets are sent one by one to the user equipment via a secure data transmission channel, including:

[0173] The order in which the image packets, after being formatted and encapsulated, are sent is determined according to the order in which they are split.

[0174] The image packet that has not yet been sent to the user equipment and has the highest priority in the sending order is identified as the target packet and sent to the user equipment.

[0175] After receiving the continued receiving instruction from the user equipment, the target packet is marked as sent, and the process returns to determine the image packet that has not been sent to the user equipment and has the highest priority in the sending order as the target packet, and then sends the target packet to the user equipment.

[0176] In one possible implementation, to improve the user's recognition of the image, the image device can be configured with an image magnification function, allowing the user to zoom in on the image and extract more image information around the parking space.

[0177] It should be noted that in practical application scenarios, after establishing a secure data transmission channel, there are multiple implementation methods for sending the image along with the parking space identifier to the user equipment. One such method is provided here as an example. Figure 7 As shown, the specific operation steps are as follows:

[0178] Step 71: The vehicle-mounted system compresses and encodes the image, and uses a preset packet splitting algorithm to split the compressed image data packet into multiple image packets. The order of splitting is determined as the order in which the image packets are sent. This triggers step 72.

[0179] Step 72: The vehicle-mounted system adds frame headers to the parking space identifiers and each image packet based on a preset data transmission protocol, and encapsulates them according to a preset encapsulation format. This triggers step 73.

[0180] Step 73: The vehicle-mounted system sends the encapsulated parking space identifier and the highest priority, yet unsent, encapsulated target image packet to the user equipment. This triggers step 74.

[0181] Step 74: The user equipment receives the encapsulated parking space identifier and image packet, and triggers step 75.

[0182] Step 75: The user equipment sends a "continue receiving" instruction to the vehicle-mounted system, specifying that it will receive an image packet whose segmentation order is later than the target image packet. This triggers step 76.

[0183] Step 76: The vehicle system responds to the instruction to continue receiving data by determining whether there are any unsent image packets. If yes, step 73 is triggered; otherwise, step 77 is triggered.

[0184] In one possible implementation, to avoid single transmission failures caused by unforeseen factors (such as other vehicles passing by the vehicle and the user), if the vehicle does not receive a continue receiving instruction from the user equipment within a preset time period, the vehicle triggers a retransmission of the target image packet sent in step 73. If the number of retransmissions exceeds a preset limit, the target image packet is recorded as a transmission failure, and step 73 is triggered to reselect the target image packet.

[0185] Step 77: The vehicle-mounted system sends a termination command to the user equipment, triggering step 78.

[0186] Step 78: In response to the end transmission command, the user equipment parses and restores the received encapsulated parking space identifier and each image packet to obtain the parking space identifier and image, and enlarges and displays the parking space identifier and image.

[0187] It should be noted that since the images in step 71 are mostly obtained by processing and stitching images acquired by various image acquisition devices by the vehicle image acquisition system, their data volume is large, resulting in low data transmission efficiency. If the user leaves the vehicle quickly after parking, or if there are structural obstructions or signal interference in the parking lot, there is a risk that image transmission may not be completed before communication between the user equipment and the vehicle's infotainment system is interrupted. This application addresses this by configuring step 71 to split the compressed and encoded images, reducing the data volume of a single image packet, and by configuring step 73 to send the encapsulated image packets to the user equipment in batches, further reducing the data volume of a single transmission, thereby improving image transmission efficiency and avoiding the risk that image transmission may not be completed before communication between the user equipment and the vehicle's infotainment system is interrupted.

[0188] In one possible implementation, the above is as follows: Figure 7 The implementation of step 71 shown can be:

[0189] Assuming the image is an AVM image, it consists of the front image, rear image, left side image, right side image, and under-vehicle image of the vehicle.

[0190] The AVM image is then split according to the image stitching boundary of the AVM using a preset image segmentation algorithm to obtain front image packet, rear image packet, left image packet, right image packet and under-vehicle image packet.

[0191] In one possible implementation, the above is as follows: Figure 7 In step 74 shown, the images are reassembled and restored based on a preset packet assembly script to obtain the images. One possible implementation is as follows:

[0192] Using a preset image assembly script, the image packages are assembled and restored according to the markings of the image stitching boundaries of each package to obtain the image in step 71 above.

[0193] To facilitate the discussion of the above... Figure 1 The understanding of the vehicle-finding assistance method shown is explained here in conjunction with one possible implementation of this application:

[0194] like Figure 8 The diagram shown is a flowchart of an auxiliary vehicle location method. The specific operation steps are as follows:

[0195] Step S801: Dynamically monitor vehicle control signals. Then trigger step S802.

[0196] Step S802: Determine whether a parking signal is detected. If yes, trigger step S803; otherwise, trigger step S801.

[0197] Step S803: Acquire images of the area surrounding the vehicle and establish a data transmission channel with the user equipment. Then, step S804 is triggered.

[0198] Step S804: Perform text detection on the image to obtain at least one parking space information, and filter the parking space information based on the position of the parking space sign in the image and the confidence level of the parking space sign. Then trigger step S805.

[0199] Step S805: Determine the parking space type based on the remaining parking space information after filtering, and determine the image deformation information of the parking space identifier in the parking space information based on the parking space type. Then trigger step S806.

[0200] Step S806: Based on the image deformation information, rotate and correct the parking space sign image in the image, and extract text from the rotated and corrected parking space sign image to obtain the parking space sign. Then, step S807 is triggered.

[0201] Step S807: The image is unpacked to obtain multiple image packets, and the parking space identifier and the image packets are encapsulated. Step S808 is then triggered.

[0202] In step S808, the encapsulated parking space identification and image packets are sent to the user equipment through the data transmission channel.

[0203] It should be noted that in practical application scenarios, the above-mentioned... Figure 8 Steps S802 and S803 shown are as follows: Figure 1 One possible implementation of step S101 is shown above. Figure 8 Step S804 shown is as follows Figure 1 One possible implementation of step S102 shown above. Figure 8 Step S805 shown is as follows Figure 1 One possible implementation of steps S103 and S104 shown above.Figure 8 Step S806 shown is as follows Figure 1 One possible implementation of step S105 is shown above. Figure 8 Step S808 shown is as follows Figure 1 One possible implementation of step S106 shown.

[0204] The second aspect of this application provides a vehicle-finding assistance system, such as... Figure 9 As shown, the vehicle location assistance system includes:

[0205] Image acquisition module 901 is used to acquire images of the area around the vehicle when a parking signal is detected;

[0206] The filtering module 902 is used to perform text detection on the image to obtain at least one parking space information, which includes: the position of the parking space sign in the image and the confidence level of the parking space sign; and to filter the parking space information based at least on the position and confidence level.

[0207] The parking space type determination module 903 is used to determine the parking space type based on the remaining parking space information after filtering. The parking space type includes perpendicular parking spaces and lateral parking spaces.

[0208] The information adding module 904 is used to add image deformation information to the position of the parking space identifier in the image of the remaining parking space information after filtering, based on the parking space type.

[0209] The text extraction module 905 is used to extract text from images containing image deformation information to obtain parking space markings.

[0210] The first data transmission module 906 is used to send parking space identifiers to user equipment.

[0211] In one possible implementation, the parking space type determination module 903 is configured as follows:

[0212] Information on each parking space:

[0213] Obtain the x and y coordinates of the center point of the location in a preset coordinate system. The origin of the preset coordinate system is the center point of the vehicle. The y-axis of the preset coordinate system is the straight line determined by the center point of the front of the vehicle, the center point of the vehicle, and the center point of the rear of the vehicle. The x-axis of the preset coordinate system is the straight line that passes through the center point of the vehicle and is perpendicular to the y-axis.

[0214] If the minimum distance between the horizontal axis and the vertical axis is not greater than the first preset threshold, and the minimum distance between the vertical axis and the horizontal axis is greater than the second preset threshold, the parking space type is determined to be a perpendicular parking space.

[0215] If the minimum distance between the horizontal axis and the vertical axis is greater than the third preset threshold, and the minimum distance between the vertical axis and the horizontal axis is not greater than the fourth preset threshold, the parking space type is determined to be a side parking space.

[0216] In one possible implementation, the aforementioned information adding module 904 is configured as follows:

[0217] For the remaining parking spaces after filtering:

[0218] The relative angle of each character in the parking space identifier relative to the vehicle is identified in the parking space information.

[0219] Calculate the angle difference between each relative angle and the preset character angle corresponding to the parking space type, and establish the correspondence between the angle difference and the character. The angle difference is an angle vector with a deviation direction.

[0220] The angle difference and its correspondence are used as image deformation information and added to the position of the parking space sign in the image.

[0221] In one possible implementation, the text extraction module 905 described above is configured as follows:

[0222] Based on the angle difference in the image deformation information, the characters of the parking space sign at the location where the image deformation information is added are rotated and corrected so that the relative angle of each character to the vehicle is equal to the preset character angle. The image of the rotated and corrected position is then extracted to obtain the parking space sign image.

[0223] Text extraction is performed on the parking space sign image to obtain the parking space sign.

[0224] In one possible implementation, the filtering module 902 described above is configured to perform text detection on the image to obtain at least one parking space information as follows:

[0225] Identify the position of each string image in the current parking space, the minimum occupied area of ​​the string image, and the confidence level of the string image in the image, and determine the position, minimum occupied area, and confidence level of the same string image as a parking space information.

[0226] In one possible implementation, the filtering module 902 described above is configured to filter parking space information based at least on location and confidence level as follows:

[0227] Remove all parking space information except for the target parking space information. The target parking space information is: the parking space information whose confidence level is within the preset confidence level range, whose minimum occupied area size is not less than the first preset size threshold, whose location is within the preset spatial area range, and whose location is the smallest distance from the center of the vehicle. The preset spatial area range is the area range composed of multiple layout areas of the parking space identifier in the parking space.

[0228] In one possible implementation, the above is as follows: Figure 9 The vehicle location assistance system shown also includes:

[0229] The second data transmission module is used to send the image along with the parking space sign to the user equipment.

[0230] In one possible implementation, the aforementioned second data sending module is specifically configured as follows:

[0231] The image is compressed and encoded to obtain a compressed image data packet;

[0232] The compressed image data packet is split into multiple image packets, and the parking space identifier and each image packet are format-encapsulated; the format-encapsulated parking space identifier and each format-encapsulated image packet are sent to the user equipment.

[0233] In one possible implementation, the second data sending module described above is configured to send the formatted parking space identifiers and each image packet to the user equipment as follows:

[0234] Establish a secure data transmission channel with user equipment;

[0235] The formatted parking space identifiers are sent to the user equipment through a secure data transmission channel, so that the user equipment can parse and display the formatted parking space identifiers.

[0236] Each formatted image packet is sent to the user equipment one by one through a secure data transmission channel. After all formatted image packets have been sent to the user equipment, a termination command is sent to the user equipment so that the user equipment can respond to the termination command by parsing and restoring the received formatted image packets to obtain the image.

[0237] In one possible implementation, the second data sending module described above is configured to send the format-encapsulated image packets one by one to the user equipment through a secure data transmission channel as follows:

[0238] The order in which the image packets, after being formatted and encapsulated, are sent is determined according to the order in which they are split.

[0239] The image packet that has not yet been sent to the user equipment and has the highest priority in the sending order is identified as the target packet and sent to the user equipment.

[0240] After receiving the continued receiving instruction from the user equipment, the target packet is marked as sent, and the process returns to determine the image packet that has not been sent to the user equipment and has the highest priority in the sending order as the target packet, and then sends the target packet to the user equipment.

[0241] In one possible implementation, the above is as follows: Figure 9 The vehicle location assistance system shown also includes:

[0242] The second data transmission module is used to send the image along with the parking space sign to the user equipment.

[0243] A third aspect of this application provides an electronic device, including at least one processor and a memory connected to the processor, wherein:

[0244] Memory is used to store computer programs;

[0245] The processor is used to execute computer programs to enable electronic devices to implement the vehicle-finding method of the first aspect or any implementation thereof.

[0246] In one possible implementation, a schematic diagram of the electronic device provided in the third aspect of this application is shown below. Figure 10 As shown, the electronic devices in the embodiments of this application may include, but are not limited to, fixed terminals such as mobile phones, laptops, PDAs (personal digital assistants), PADs (tablet computers), desktop computers, etc. Figure 10 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0247] like Figure 10 As shown, the electronic device may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 1001, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1008 into a random access memory (RAM) 1003. When the electronic device is powered on, the RAM 1003 also stores various programs and data required for the operation of the electronic device. The processing unit 1001, ROM 1002, and RAM 1003 are interconnected via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0248] Typically, the following devices can be connected to the I / O interface 1005: input devices 1006 including, for example, a touchscreen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 1007 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1008 including, for example, memory card, hard disk, etc.; and communication devices 1009. Communication device 1009 allows electronic devices to exchange data via wireless or wired communication with other devices. Although Figure 10 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown. More or fewer devices may be implemented or have alternatively.

[0249] This application also provides a computer program product including computer-readable instructions, which, when executed on an electronic device, cause the electronic device to implement any of the vehicle-finding methods provided in this application.

[0250] It should also be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. In addition, in the device embodiment drawings provided in this application, the connection relationship between modules indicates that they have a communication connection, which can be implemented as one or more communication buses or signal lines.

[0251] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware, or it can be implemented by special-purpose hardware including application-specific integrated circuits, special-purpose CPUs, special-purpose memory, special-purpose components, etc. Generally, any function performed by a computer program can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be diverse, such as analog circuits, digital circuits, or special-purpose circuits. However, for this application, software program implementation is more often the preferred implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium, such as a computer floppy disk, USB flash drive, mobile hard disk, ROM, RAM, magnetic disk, or optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, training equipment, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0252] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product.

[0253] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, training device, or data center to another website, computer, training device, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)).

Claims

1. A vehicle-finding assistance method, characterized in that, include: When a stop signal is detected, images of the area around the vehicle are acquired; Text detection is performed on the image to obtain at least one parking space information, the parking space information including: the position of the parking space identifier in the image and the confidence level of the parking space identifier; the parking space information is filtered based at least on the position and the confidence level; Based on the remaining parking space information after filtering, the parking space type is determined, including perpendicular parking spaces and lateral parking spaces; Based on the parking space type, image deformation information is added to the position of the parking space identifier in the image of the remaining parking space information after filtering; Text extraction is performed on the image containing the image deformation information to obtain the parking space identifier; The parking space identifier is sent to the user device.

2. The vehicle-finding assistance method according to claim 1, characterized in that, The step of determining the parking space type based on the remaining parking space information after filtering includes: Regarding the parking space information mentioned above: Obtain the x-coordinate and y-coordinate of the center point of the location in a preset coordinate system. The origin of the preset coordinate system is the center point of the vehicle. The y-axis of the preset coordinate system is a straight line determined by the center point of the front of the vehicle, the center point of the vehicle, and the center point of the rear of the vehicle. The x-axis of the preset coordinate system is a straight line passing through the center point of the vehicle and perpendicular to the y-axis. If the minimum distance between the horizontal axis and the vertical axis is not greater than a first preset threshold, and the minimum distance between the vertical axis and the horizontal axis is greater than a second preset threshold, then the parking space type is determined to be the vertical parking space. If the minimum distance between the horizontal axis and the vertical axis is greater than a third preset threshold, and the minimum distance between the vertical axis and the horizontal axis is not greater than a fourth preset threshold, then the parking space type is determined to be the side parking space.

3. The vehicle-finding assistance method according to claim 1, characterized in that, The step of adding image deformation information to the position of the parking space identifier in the image of the remaining parking space information after filtering, based on the parking space type, includes: For the remaining parking space information after filtering: The relative angle of each character of the parking space identifier in the parking space information to the vehicle is identified; Calculate the angle difference between each relative angle and the preset character angle corresponding to the parking space type, and establish the correspondence between the angle difference and the character. The angle difference is an angle vector with a deviation direction. The angle difference and the corresponding relationship are used as the image deformation information and added to the position of the parking space identifier in the image.

4. The vehicle-finding assistance method according to claim 3, characterized in that, The step of extracting text from the image containing the image deformation information to obtain the parking space identifier includes: Based on the angle difference in the image deformation information, each character of the parking space sign at the position where the image deformation information is added in the image is rotated and corrected so that the relative angle of each character with respect to the vehicle is equal to the preset character angle. Then, the image of the position after the rotation correction is extracted to obtain the parking space sign image. The text is extracted from the parking space sign image to obtain the parking space sign.

5. The vehicle-finding assistance method according to claim 1, characterized in that, The step of performing text detection on the image to obtain at least one parking space information includes: Identify the position of each string image in the current parking space, the minimum occupied area of ​​the string image, and the confidence level of the string image in the image, and determine the position, the minimum occupied area, and the confidence level of the same string image as a parking space information.

6. The vehicle-finding assistance method according to claim 5, characterized in that, The filtering of the parking space information based at least on the location and the confidence level includes: Remove all parking space information except for the target parking space information. The target parking space information is the parking space information that has a confidence level within a preset confidence level range, a minimum occupied area size not less than a first preset size threshold, a location within a preset spatial area range, and a location with the smallest distance from the vehicle center. The preset spatial area range is the area range composed of multiple layout areas of parking space identifiers in the parking space.

7. The vehicle-finding assistance method according to claim 1, characterized in that, The method further includes: The image is sent to the user equipment along with the parking space identifier.

8. The vehicle locator method according to claim 7, characterized in that, Sending the image along with the parking space identifier to the user equipment includes: The image is compressed and encoded to obtain a compressed image data packet; The compressed image data packet is split into multiple image packets, and the parking space identifier and each of the image packets are format-encapsulated; the format-encapsulated parking space identifier and each format-encapsulated image packet are sent to the user equipment.

9. The vehicle-finding assistance method according to claim 8, characterized in that, The step of sending the parking space identifiers and each image packet, encapsulated in the aforementioned format, to the user equipment includes: Establish a secure data transmission channel with the user equipment; The parking space identifier encapsulated in the specified format is sent to the user equipment through the secure data transmission channel, so that the user equipment can parse and display the parking space identifier encapsulated in the specified format. Each image packet encapsulated in the specified format is sent to the user equipment one by one through the secure data transmission channel. After all image packets encapsulated in the specified format have been sent to the user equipment, a termination transmission command is sent to the user equipment so that the user equipment responds to the termination transmission command by parsing and restoring the received image packets encapsulated in the specified format to obtain the image.

10. The vehicle-finding assistance method according to claim 9, characterized in that, The step of sending each image packet, encapsulated in the specified format, to the user equipment one by one through the secure data transmission channel includes: The order in which the image packets encapsulated in the specified format are sent is determined according to the order in which they are split. The image packet that has not been sent to the user equipment and has the highest priority in the sending order is identified as the target packet after being encapsulated in the specified format, and the target packet is sent to the user equipment. After receiving the continued receiving instruction from the user equipment, the target packet is marked as sent, and the process returns to execute the operation steps of determining the image packet encapsulated in the specified format that was not sent to the user equipment and has the highest priority in the sending order as the target packet, and sending the target packet to the user equipment.

11. A vehicle-finding assistance system, characterized in that, include: The image acquisition module is used to acquire images of the area around the vehicle when a stop signal is detected. A filtering module is used to perform text detection on the image to obtain at least one parking space information, the parking space information including: the position of the parking space identifier in the image and the confidence level of the parking space identifier; and to filter the parking space information based at least on the position and the confidence level. The parking space type determination module is used to determine the parking space type based on the remaining parking space information after filtering. The parking space type includes perpendicular parking spaces and lateral parking spaces. The information adding module is used to add image deformation information to the position of the parking space identifier in the image of the remaining parking space information after filtering, based on the parking space type. The text extraction module is used to extract text from the image containing the image deformation information to obtain the parking space identifier; The first data transmission module is used to send the parking space identifier to the user equipment.

12. An electronic device, characterized in that, It includes at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program to enable the electronic device to implement the vehicle-finding assistance method as described in any one of claims 1 to 10.

13. A computer storage medium, characterized in that, The storage medium carries one or more computer programs that, when executed by an electronic device, enable the electronic device to implement the vehicle-finding assistance method as described in any one of claims 1 to 10.

Images